cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: FFTW 3.3.5: Installation and Customization cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127:
cannam@127:

cannam@127: Next: , Previous: , Up: Top   [Contents][Index]

cannam@127:
cannam@127:
cannam@127: cannam@127:

10 Installation and Customization

cannam@127: cannam@127: cannam@127:

This chapter describes the installation and customization of FFTW, the cannam@127: latest version of which may be downloaded from cannam@127: the FFTW home page. cannam@127:

cannam@127:

In principle, FFTW should work on any system with an ANSI C compiler cannam@127: (gcc is fine). However, planner time is drastically reduced if cannam@127: FFTW can exploit a hardware cycle counter; FFTW comes with cycle-counter cannam@127: support for all modern general-purpose CPUs, but you may need to add a cannam@127: couple of lines of code if your compiler is not yet supported cannam@127: (see Cycle Counters). (On Unix, there will be a warning at the end cannam@127: of the configure output if no cycle counter is found.) cannam@127: cannam@127: cannam@127: cannam@127:

cannam@127: cannam@127:

Installation of FFTW is simplest if you have a Unix or a GNU system, cannam@127: such as GNU/Linux, and we describe this case in the first section below, cannam@127: including the use of special configuration options to e.g. install cannam@127: different precisions or exploit optimizations for particular cannam@127: architectures (e.g. SIMD). Compilation on non-Unix systems is a more cannam@127: manual process, but we outline the procedure in the second section. It cannam@127: is also likely that pre-compiled binaries will be available for popular cannam@127: systems. cannam@127:

cannam@127:

Finally, we describe how you can customize FFTW for particular needs by cannam@127: generating codelets for fast transforms of sizes not supported cannam@127: efficiently by the standard FFTW distribution. cannam@127: cannam@127:

cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: cannam@127:
cannam@127:
cannam@127:

cannam@127: Next: , Previous: , Up: Top   [Contents][Index]

cannam@127:
cannam@127: cannam@127: cannam@127: cannam@127: cannam@127: